Sustainability teams are stretched to their limits. They handle everything from reporting and compliance to developing responsible sourcing strategies and gathering data from countless sources. Even at major brands, these teams are often small, spending more time buried in spreadsheets than driving the meaningful sustainability strategies they originally set out to implement. It’s no wonder that the promise of highly efficient AI is so appealing.
Industry experts suggest AI can relieve pressure on both sustainability teams and their supply chain partners by automating environmental reporting, improving data quality, reformatting information for different needs, and verifying traceability. AI also offers potential efficiency gains within the supply chain itself, through smarter material use and more accurate demand forecasting. “AI is definitely helping sustainability teams with reporting, allowing them to focus more on strategic programs rather than compliance,” says Annie Agle, vice president of impact and sustainability at outdoor brand Cotopaxi.
However, there’s a caveat. The true environmental impact of using AI at an organizational level remains largely unclear. Without careful monitoring, brands could unintentionally increase their carbon footprint through the very technology meant to reduce it.
“The benefits are clear, but we also know there are negative impacts that aren’t fully understood yet,” Agle notes. “We don’t yet know how that digital footprint affects our greenhouse gas measurements.”
To learn how sustainability teams are integrating AI and addressing these uncertainties, Vogue Business spoke with a selection of brands of varying sizes and market segments. While some have been using AI for years, others are still experimenting, but all agree AI will play a role in their future operations.
How Sustainability Teams Are Using AI
H&M reports using AI across its supply chain, logistics, marketing, sales, and customer experience. The brand says AI supports its goal of producing only what it sells by optimizing production quantities, sales locations, and timing. “This has positive effects on resource use, inventory, raw materials, and emissions,” H&M explains.
Luxury group Kering, which owns brands like Gucci and Balenciaga, appointed Pierre Houlès as chief digital, AI and IT officer in March 2026. The company has been implementing AI across select houses for several years. Like H&M, Kering uses analytical AI to forecast demand and optimize inventory levels for each product, according to chief sustainability and institutional affairs officer Marie-Claire Daveu. Similar to Cotopaxi, it also employs AI to automate and enhance reporting reliability, using tools that automatically collect and intelligently correct data from Kering sites. At the product development level, Kering’s Italy-based Material Innovation Lab—which researches more sustainable materials—created an eco-design AI agent to provide technical guidance to design teams. This helps bridge often separated design and sustainability teams, ensuring one group’s output doesn’t undermine the other’s goals.
While H&M and Kering have integrated AI into core operations over several years, other brands are using it primarily to ease workloads. “When it comes to AI, we’re still in the early stages of exploration,” said Everlane CEO Alfred Chang in a statement to Vogue Business. “We’re focused on how it might support day-to-day internal processes and save time.”H&M SS26.
Photo: Courtesy of H&M Group
Agle notes that her small team uses AI to translate raw supplier data into greenhouse gas (GHG) measurements, reformat data for various reports, and create visualizations to help the wider team understand company impact. “It really does help with productivity. I definitely think companies that don’t adopt AI are going to struggle to compete commercially,” she says.
Maximilien Abadie, deputy CEO of French technology company Lectra, agrees. Lectra offers AI solutions that help fashion brands meet sustainability goals, such as verifying traceability data and optimizing fabric cutting to reduce over-ordering. But Abadie believes the greatest opportunity lies in shortening time to market. “The challenge for any fashion company is how to stay competitive in a world full of daily disruptions and uncertainties, where you can’t predict what will happen in three or six months. You need to be present at the right time with the right product, at the right price, in the right quantities, for the right consumer.”
Maximilien Abadie, deputy CEO of French technology company Lectra.
Photo: Courtesy of Lectra
While AI is being used to cut waste and optimize supply chains, it is also widely deployed for commercial and creative purposes. “Since 2018, we have developed more than 15 internal machine learning platforms, all built to enhance creativity, efficiency, and customer service,” says Jordi Alex, chief information technology officer at Spanish retailer Mango.
According to Alex, Mango developed most of these platforms in-house. This includes its AI assistant Iris, which handles over 7.5 million customer inquiries per year. Another tool, Gaudi, uses customer browsing and buying data to generate personalized product recommendations. Mango also uses generative AI to create campaign and collection imagery. These extensive applications beyond sustainability increase the urgency for brands to examine AI’s underlying environmental impacts.
An AI-generated campaign image for Mango Teen.
Photo: Courtesy of Mango and generated by AI
Unknown Impacts
It is widely acknowledged that AI growth has increased demand for energy and water, the latter used for cooling data centers. A 2025 paper in Nature stated that, at current growth rates, AI servers in the U.S. alone would generate 24–44 million metric tons of carbon dioxide (CO₂)—equivalent to adding 5–10 million cars to U.S. roads. Their annual water footprint ranges from 731 to 1,125 million cubic meters, a concern compounded by many data centers being located in some of the world’s driest regions. In April 2025, the International Energy Agency (IEA) reported that data centers accounted for 1.5% of global energy consumption in 2024, a figure expected to rise to 3% by 2030.
While data centers support more than just AI, AI workloads are projected to represent half of all data center capacity by 2030, highlighting its significant and growing energy demand. U.S. AI company Anthropic has acknowledged that the grid capacity and data center growth required for its technology will increase electricity prices for consumers and has stated it will cover related infrastructure costs. The company did not respond to requests for comment.
While broader projections can be made, calculating the impact of AI at an organizational level is much trickier. “It’s going to be very complicated to understand the impacts of AI as a non-AI company. When it’s your service and you own the data centers, you can count the electricity and water usage needed for cooling,” says Cotopaxi’s Agle. But for a fashion brand that doesn’t control all the underlying infrastructure, assessing the full footprint remains a challenge.It is extremely difficult to arrive at an exact figure for how every single employee is using AI, whether they are writing emails with ChatGPT or gathering data with internal tools. This calculation also depends on AI companies measuring and validating their own impacts so that third parties can use the data. However, major players like OpenAI, Perplexity, and Anthropic have not publicly disclosed any emissions data, and none responded to requests for comment.
“It’s going to be very complicated for a non-AI company to understand the impacts of AI. When it’s your service and you own the data centers, you can count the electricity and water needed for cooling,” says one expert.
Kering states that it closely monitors the environmental impact of AI across its IT activities, including data centers and cloud usage, noting that IT represents less than 2% of its total carbon footprint. Among the companies contacted, Kering was the only one to share an outline of its mitigation strategy: prioritizing simple, resource-efficient models and working with technology partners to support decarbonization and improve carbon reporting transparency.
Kering did not specify the source of its data, and currently, there are no standardized metrics. To enable comparable assessments in the future, the Coalition for Sustainable AI—initiated by the French government, the UN Environment Programme (UNEP), and the International Telecommunications Union (ITU)—is calling for global standardization. As a step toward this, the ITU released guidelines in February 2026 for assessing and minimizing the environmental impact of AI systems. Additionally, in January 2026, the Taskforce on Nature-related Financial Disclosures (TNFD) released draft sector guidance for the technology and communications sector to help organizations assess their nature-related dependencies, impacts, risks, and opportunities. These are initial steps toward a standardized system, similar to those used for reporting on production or transportation impacts.
Until robust frameworks are established and widely adopted, brands must find their own way forward. In the absence of global standards for calculating AI’s footprint, Cotopaxi plans to use rough estimates to understand its impact. “Step one is understanding our usage. What contracts do we have with an AI provider? What kind of usage are we seeing from employees? If we have a certain number of employees using a tool like Claude, and Claude reports a set of emissions, what percentage of those do we own? It’s going to be challenging to measure, but we need to try,” says Agle of Cotopaxi.
From Measuring to Mitigating
For Cotopaxi, measuring impact is the first step in creating a responsible AI policy. Some brands already have such policies in place. H&M developed a responsible AI framework in 2018, while AI adoption at Mango is led by a steering committee. “Through centralized governance, training, and programs like AI Champions—internal ambassadors who support adoption and share best practices—we ensure AI is adopted correctly and helps our people realize their potential,” says a Mango representative. However, many existing policies focus on the ethics of AI, such as safety, anti-bias, and transparency, rather than environmental factors.
The French Ministry of Ecological Transition’s general framework for frugal AI use aims to fill that gap by offering guidance on best practices and how to move from…From measurement to mitigation, the ministry suggests brands consider several steps. They can question whether AI is truly needed for specific tasks, schedule AI training during times when data centers rely on renewable energy, use precise, high-quality training data to reduce computing demands, and explore whether a less energy-intensive AI model could be effective.
While large, multi-purpose AI platforms powered by data centers are common, alternatives exist. UK-based Literal Labs, a spinout from Newcastle University, focuses on creating efficient AI models that don’t require expensive, specialized Graphics Processing Units (GPUs). Instead, their models run on smaller, cheaper, and more efficient chips found in everyday devices like TV remotes or microwaves. They achieve this by replacing the complex neural networks behind systems like ChatGPT with simpler “if/then” logic-based networks, which demand far less computing power.
According to Daniel Dykes, the company’s chief product officer, this approach shows no trade-offs in capability. Both neural networks and logic-based networks can perform deep learning to handle complex data. In a test for a food company’s demand forecasting, their logic-based network was 7% more accurate than a leading neural network competitor.
Dykes notes, “If an algorithm doesn’t require specialized hardware, a data center, or a new power plant, you’ve solved many of the problems posed by current AI.” The company, which operates in the EMEA region and sectors like water utilities, claims its training process is over 50 times more energy efficient and 54 times faster than equivalent neural networks.
Another UK innovator, DeepGate, is developing AI that runs on smaller chips, efficiently detecting “wake words” (like “Hey Siri”) and classifying images. While not a full replacement for more demanding systems, it shows that many specialized tasks can run on cheaper, energy-efficient hardware instead of GPUs. Examples include inspecting material quality or spotting irregularities in documents and supply chain data—targeted, specific applications.
DeepGate co-founder Luke Taylor explains that these simpler systems can complement larger ones, handling specialized tasks only when needed rather than running continuously, which wastes energy.
Solutions are emerging, but to use them strategically, brands must first understand AI’s full impact to ensure it benefits sustainability. As Lectra’s Abadie points out, “It doesn’t make sense to improve something in one area while increasing pollution and CO₂ emissions elsewhere.” He believes greater awareness will lead to shorter development cycles and better AI. “As we progress, more people will realize that environmental impact must be considered from the start. If AI creates more harm than good, it’s not worth investing in.”
Frequently Asked Questions
FAQs Will AI Make Fashion More Sustainable or Less
BeginnerLevel Questions
1 What does sustainable fashion even mean
Sustainable fashion refers to clothing that is designed manufactured distributed and used in ways that are environmentally friendly and socially responsible It aims to minimize harm to the planet and ensure fair conditions for workers
2 How can AI help fashion at all
AI can help in many ways like predicting trends to reduce overproduction designing clothes with less waste optimizing supply chains to save energy and helping customers find items theyll love and keep longer
3 So is AI basically a good thing for making fashion greener
It has the potential to be a powerful tool for good but its not automatic The outcome depends on how companies choose to use it If used to optimize for sustainability it can help If used solely to drive more consumption faster it could make things worse
4 Can AI really design clothes
Yes AI can generate new clothing designs based on specific parameters This can help designers experiment and create patterns that minimize fabric waste from the very beginning
Advanced Practical Questions
5 Whats the biggest potential benefit of AI for sustainability
Reducing Overproduction The fashion industrys biggest environmental sin is making billions of unsold items that end up in landfills or burned AIs advanced demand forecasting can help brands make closer to what they will actually sell drastically cutting waste
6 What are the specific risks or downsides of using AI in fashion
Increased Speeds AI could enable ultrafast hypertrends encouraging even more disposable consumption
High Energy Use Training and running complex AI models requires significant computing power which has a large carbon footprint
Greenwashing Brands might use AI as a buzzword to appear sustainable without making meaningful systemic changes
Job Displacement It could automate roles in design forecasting and manufacturing raising social sustainability concerns
7 Are there realworld examples of AI helping right now
Yes Companies are using AI for
OnDemand Manufacturing Only producing an item once its ordered